一种基于二维算法的新颖的多目标光纤光谱数据处理流程

A Novel Spectral Data Processing Procedure on Multi-Object Fiber Spectral Data Based on 2-D Algorithms

郭守敬望远镜(Large Sky Area Multi-Object Fiber Spectroscopic Telescope,LAMOST)、斯隆数字巡天(Sloan Digital Sky Survey,SDSS)、英澳望远镜(AngloAustralia Telescope,AAT)等大多数多目标光纤光谱望远镜现用的数据处理流程都是基于一维算法的.以LAMOST为例提出多目标光纤光谱数据处理流程方法.在LAMOST现用数据处理流程中,在预处理过程之后,通过基于一维模型的抽谱算法从二维观测目标光谱数据中得到一维抽谱结果作为中间数据.后续的处理步骤都基于一维模型的算法.然而,这种数据处理流程不符合观测光谱的形成机理.因此,在每个步骤中都引入了不可忽略的误差.为了解决这一问题,提出了一种还未被用于LAMOST及其他望远镜数据处理系统的新颖的数据处理流程.重新设计安排了各个数据处理模块的顺序,各关键步骤算法都是基于二维模型的.核心算法将详细论述.此外,列出了部分实验结果来证明二维算法的有效性和优越性.

The data processing procedures currently used on most multi-object fiber spectroscopic telescopes, such as Large Sky Area Multi-Object Fiber Spectroscopic Tele- scope （LAMOST）, the Sloan Digital Sky Survey （SDSS）, the Anglo-Australia Telescope （AAT）, etc., are based on one-dimensional （l-D） algorithms. In this paper, LAMOST is taken as an example to display the proposed multi-object fiber spectral data processing procedure. In the using processing procedure on LAMOST, after the pretreatment pro- cess, the two-dimensional （2-D） observed raw data are extracted into 1-D intermediate data simply based on 1-D model. Then the subsequent key steps are all done by 1-D algorithms. However, this processing procedure is not in accord with the formation mechanism of the observed spectra. Therefore, it brings a considerable error in each step. To solve the problem, we propose a novel processing procedure that has not been used on LAMOST or other telescopes. The modules of the procedure are reordered, and the main steps are all based on 2-D algorithms. The principles of the core algorithms are explained in detail. Besides, some partial experimental results are shown to prove the effectiveness and superiority of the 2-D algorithms.